Where in perlre, the documentation for Perl Regular Expressions did you find that \< means "start of word"? I only find one mention of \<, and that's in the sentence:

So anything that looks like \\, \(, \), \<, \>, \{, or \} is always interpreted as a literal character, not a metacharacter.

.. and that seems to me that it specifically rules out that \< has any meaning beyond a literal <.

#!/usr/bin/perl
use strict;
use warnings;

#885479

my $string = " cat mat rat pat\n"; 
if ($string =~ /\bcat/) {
     print "yes\n"; 
     } else {
     print "no\n"; 
}
[download]

Note also the omission of your exit 0; and substitution of use warnings (scoped to the current script whereas  - w may kick out warnings about any modules you might use in some more complex program.

However, for future posts, please enclose code (and data) in <c>...</c> tags. See Markup in the Monastery.

Updated with MRE citation.

Perl uses \b to signify a place between a "word" and a "non-word" character, which is somewhat different from vi's notion. And Perl doesn't distinguish between "begin word" and "end word".

$string =~ /\bcat\b/
$string =~ /\Wcat\W/
$string =~ /\scat\s/
[download]

Update: another interesting read is 7 Stages of Regex Users...